Vit Facial Expression Recognition

🚀 vit-Facial-Expression-Recognition

This model is a fine - tuned Vision Transformer for facial emotion recognition, leveraging multiple datasets to achieve high - accuracy emotion classification.

This model is a fine - tuned version of [google/vit - base - patch16 - 224 - in21k](https://huggingface.co/google/vit - base - patch16 - 224 - in21k) on the FER 2013, MMI Facial Expression Database, and [AffectNet dataset](https://www.kaggle.com/datasets/noamsegal/affectnet - training - data) datasets. It achieves the following results on the evaluation set:

• Loss: 0.4503
• Accuracy: 0.8434

🚀 Quick Start

This model can be used for facial emotion recognition tasks. First, ensure you have the necessary libraries installed. You can then load the model and perform predictions on facial images.

✨ Features

• Fine - tuned for Emotion Recognition: Specifically optimized for the task of classifying facial emotions.
• Multi - dataset Training: Trained on a combination of diverse datasets, enhancing its generalization ability.
• High Accuracy: Achieves an accuracy of 0.8434 on the evaluation set.

📚 Documentation

Model description

The vit - face - expression model is a Vision Transformer fine - tuned for the task of facial emotion recognition.

It is trained on the FER2013, MMI facial Expression, and AffectNet datasets, which consist of facial images categorized into seven different emotions:

• Angry
• Disgust
• Fear
• Happy
• Sad
• Surprise
• Neutral

Data Preprocessing

The input images are preprocessed before being fed into the model. The preprocessing steps include:

• Resizing: Images are resized to the specified input size.
• Normalization: Pixel values are normalized to a specific range.
• Data Augmentation: Random transformations such as rotations, flips, and zooms are applied to augment the training dataset.

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 3e - 05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9, 0.999) and epsilon = 1e - 08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 3

Training results

Framework versions

• Transformers 4.36.0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.15.0